The pressure for an attack on Iran is building. Media reports suggest that the Obama administration is under pressure to take action and may even be considering action itself, and Foreign Affairs published an article in its opening 2012 issue by nuclear expert Matthew Kroenig forthrightly stating that it is “time to attack Iran.” Many argue that strikes against Iran’s nuclear program are the only responsible course. Defense Secretary Leon Panetta declared on December 18 that if the United States receives “intelligence that [the Iranians] are proceeding with developing a nuclear weapon, then we will take whatever steps necessary to stop it.” And former White House Middle East chief Dennis Ross suggested in a December 23 op-ed that, if pressure fails, the military option is the only sensible recourse. The United States cannot live with a nuclear Iran, he argues.

The reality, however, is that attacking Iran without provocation is a dangerous course. The arguments for avoiding military strikes are well known: deterrence, while neither easy nor cheap, can work; the costs of likely Iranian retaliation outweigh the likely benefits, perhaps markedly; and the United States (and its allies) have considered preventive attacks against adversary nuclear programs before, thought the better of it and come out tolerably.

But perhaps the most important argument against attacking Iran has received less attention. That is that none of the attack proponents can give a sensible answer to the question General David Petraeus posed at the beginning of the Iraq war: “How does this end?” Kroenig and other advocates for war note, correctly, that a strike against Iran could do substantial damage to Iran’s program. But they fail to explain how the United States will prevent Iran from simply restarting its program, this time in deadly earnest. Moreover, they don’t explain why such strikes won’t contribute to the immediate rallying of the Iranian people around the otherwise reviled regime.

Yet these questions need answers. If the United States strikes Iran and fails to follow up, Washington might delay Iran’s acquisition of a nuclear weapon, but it will give Tehran every incentive to reboot the program with greater vigor. This is not mere theorizing. After Israel’s much-lauded 1981 strike on Iraq’s Osirak plutonium-reprocessing facility, Baghdad redoubled its efforts to obtain a weapon, only failing to achieve its goal because its pursuit was cut short by Desert Storm. And the dogs that didn’t bark are perhaps even more illuminating; Washington considered preventive attacks on the Soviet, Chinese and North Korean nuclear programs but decided not to follow through. Such strikes were believed to be costly as well as ultimately futile and counterproductive.

The basic question is: How do attack advocates propose to stop the Iranian nuclear program if Tehran refuses to roll over after one round of attacks? There are two logical responses to this question. One is regime change, presumably through invasion. But there are significant downsides to invasion, not least that such a war would likely prove protracted and costly. Attack advocates such as Kroenig effectively concede that the American people are unlikely to support this course.

The other is that the United States should be prepared to conduct repeated strikes over a long period of time to ensure the Iranian nuclear program is kept down. Unsurprisingly, Kroenig and others shy away from this answer, as it is a recipe for perpetual war. The cost in lives, resources and America’s international reputation would be formidable, especially if done without diplomatic cover and international support that probably wouldn’t be forthcoming. Yet, even under the most favorable conditions in which Iranian retaliation stayed limited and international support was forthcoming, a long-term, limited-strike campaign might not work at a level of effort and damage in line with U.S. aims. Regular U.S. strikes on North Vietnam over a period of seven years under highly favorable international conditions failed woefully either to convince Hanoi to change its fundamental strategy or substantially degrade the communist war effort. The North Vietnamese resolutely repaired bridges, depots and roads. More recently, limited allied air strikes against Iraq in the 1990s didn’t force Saddam Hussein’s compliance with UN Security Council Resolutions. Moreover, as happened in Vietnam, such strikes likely would become more difficult over time. The Russians, for example, have refrained, thanks to Western diplomacy, from selling Iran advanced long-range surface-to-air missiles that could make strikes more difficult. They probably wouldn’t be so forbearing if strikes were conducted without their prior and explicit approval, which Moscow isn’t likely to give.

But the primary challenges to U.S. striking power would be political and stem from the difficulties in sustaining a coalition supportive of such measures over time. Recall that the consensus behind Clinton-era strikes on Iraq fell apart as the policy continued with no end in sight. Ultimately, such a campaign would be unlikely to succeed absent a fundamental change in the perception of the Iranian government. Failing such a change, a perpetual-strike campaign would leave us in a worse situation. Tehran would have an added incentive to obtain a nuclear capability to strengthen its deterrent against attack. U.S. military power would be shown to be of limited value. And an international coalition against Iran’s nuclear program would fall apart. Finally, if the Iranian regime did succeed in acquiring a nuclear capability even after a sustained U.S. attack, it would be sure to be even more paranoid and on edge, with deeply worrying implications for how it would posture its nuclear arsenal. Since Tehran would reasonably believe the United States and other powers were aiming to take military action to deprive it of its nuclear weapons, Iran would have a strong incentive to take destabilizing steps to ensure those weapons were protected and, ultimately, that they could be used. No one should want Tehran delegating nuclear-launch authority to lower-level commanders, for example, or prepositioning nuclear weapons with Hezbollah, precisely the kind of steps Iran might choose to take in such circumstances.

Many attack proponents argue that U.S. military strikes against Iran would embolden resistance to the government. If this were true, there would be much to recommend them. But nothing is more likely to spur support for the Islamic Republic regime than repeated air strikes by the United States or under its auspices. Large-scale bombing campaigns didn’t break support for North Vietnamese or North Korean regimes, or for the German or Japanese governments during World War II. Rather, they hardened support for them. Iran showed gruesome fortitude and chilling cohesion during the Iran-Iraq War, when it sent waves of youths into mass infantry attacks, and nothing in Persian history or today’s Iran gives reason to think Iran would do anything but rally around the flag of the Islamic Republic when under attack.

We acknowledge, of course, that containing a nuclear Iran would be costly and risky. The United States would need to be strong, resolute and even fearsome in demonstrating to Iran the costs of aggression and assuring U.S. allies that staying the course with Washington represents a prudent strategy. Yet attacking Iran means rallying ordinary Iranians to a regime they dislike and many despise, and it risks a wider war in the region. And it would alienate key international actors, such as Russia, whose support would be necessary to ensure that an effective sanctions policy could work over time. And an attack would do all this without even providing a reasonable and plausible answer to the ultimate question Americans want answered before the United States goes to war: How does this end? Stealthy air strikes and massive earth-penetrating bombs are only tools, not answers. The United States cannot responsibly attack Iran and leave it at that, simply hoping for the best. A firm and resolute containment may be costly and risky, but it is a lot better than that. It’s probably best not to start down a road that has no end in sight.

Elbridge Colby is a defense analyst who previously served in a number of U.S. Government positions, including as an expert advisor to the Congressional Strategic Posture, for which he prepared a study on the question of deterring a nuclear-armed Iran. Austin Long is an assistant professor at Columbia University and author of a detailed assessment of the merits and implications of an attack on Iranian nuclear facilities.